From Builder Skills
Enables autonomous progress by making reasonable assumptions, avoiding routine clarifications, and logging decisions. Stops only for true blockers like missing secrets or destructive actions.
How this skill is triggered — by the user, by Claude, or both
Slash command
/builder-skills:plow-aheadThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Proceed through ordinary ambiguity. Make reasonable assumptions, keep momentum,
Proceed through ordinary ambiguity. Make reasonable assumptions, keep momentum, validate as you go, and make the final recap strong enough that the user can see what decisions were made while they were away.
Treat the user's instruction as permission to continue through normal uncertainty:
Stop and ask only for true blockers:
If blocked, leave a self-contained handoff: what was done, what blocks progress, what exact input is needed, and the next command or file to inspect.
When choosing without the user:
Maintain a lightweight decision log while working. It can live in notes, the plan, or your final answer, but do not create a new repo artifact unless the task needs one.
End with a recap that makes autonomous decisions auditable:
Goal
- What you completed.
Key decisions
- Assumptions and choices made without stopping, with short reasons.
Changes
- Files, behavior, docs, or configuration changed.
Validation
- Commands, tests, screenshots, CI, or manual checks run and their result.
Remaining risk
- Anything not verified, deferred, or blocked.
Keep the recap factual. Do not hide uncertainty, skipped validation, or judgment calls.
npx claudepluginhub builderio/skillsGuides AI agents on when to ask clarifying questions vs. proceed autonomously using checklists and decision matrices to reduce unnecessary interaction rounds.
Activates full autonomous execution mode, preventing early stops until all todo tasks are verified complete. Ideal for multi-step builds, test fixes, refactors, and migrations.
Auto-loop execution workflow with quality gates. Use when starting any non-trivial implementation task. Provides automatic task decomposition, code implementation, testing (L1-L4), and iterative quality gates until completion. Invoke with /autoworker.